v1.2.8

# ⚠️ Deprecated

## 🚫 Gaze (L2CS-Net / MediaPipe) detection

Gaze detection — the `roboflow_core/gaze@v1` workflow block, the `POST /gaze/gaze_detection` HTTP route, and the `InferenceHTTPClient.detect_gazes()` / `detect_gazes_async()` SDK helpers — has been **deprecated** in this release. Calling any of these now raises a new `FeatureDeprecatedError` that surfaces as **HTTP 410 Gone** with `error_type: "FeatureDeprecatedError"`.

The Gaze pipeline was built on top of **MediaPipe**, which has dropped support for parts of the hardware matrix Roboflow ships against (notably some Linux/aarch64 and Jetson configurations no longer have compatible wheels), and a transitive `protobuf` [CVE2026-0994](https://nvd.nist.gov/vuln/detail/CVE-2026-0994) meant we could no longer carry the dependency alongside the rest of the inference stack.

```python
# Old (still loads, returns 410 at execute time):
from inference_sdk import InferenceHTTPClient
client = InferenceHTTPClient(api_url="...", api_key="...")
client.detect_gazes(inference_input="image.jpg")
# → raises inference_sdk.http.errors.FeatureDeprecatedError
```

> [!IMPORTANT]
> **The last release supporting Gaze is `v1.2.7`.** If you currently rely on Gaze detection locally — pin to `inference==1.2.7` (or the matching Docker image tag) as a short-term bridge while you plan ahead, keeping in mind that vulnerability exists in the build.

> [!NOTE]
> The `POST /gaze/gaze_detection` route remains registered as a **410-Gone stub** through end of Q2 2026 so existing client integrations get a structured error rather than a 404. Set `CORE_MODEL_GAZE_ENABLED=False` to disable it immediately.

If Gaze is important to your workflow and you'd like to discuss bringing it back paired with a different face detector, reach out at **support@roboflow.com** — we're happy to chat.

PR: https://github.com/roboflow/inference/pull/2334

---

# 💪 Added

## 🖼️ Image Stack workflow block

A new `Image Stack` workflow block lets you accumulate frames across executions — useful for building temporal pipelines, sliding-window inference, or any flow that needs to reason over a buffer of recent images rather than a single frame.

PR: https://github.com/roboflow/inference/pull/2307

## 🤖 OpenAI-compatible LLM block for custom endpoints

A new workflow block that lets you point at **any OpenAI-compatible HTTP endpoint** — your own self-hosted vLLM/Ollama/LM Studio deployment, a private gateway, or a third-party provider that mirrors the OpenAI API. Combined with the `extra_body` follow-up, you can also pass provider-specific extensions (reasoning effort, sampling tweaks, custom routing flags) through to the upstream call.

PRs: https://github.com/roboflow/inference/pull/2309, https://github.com/roboflow/inference/pull/2313

## 🔐 Opt-in HTTPS for the inference server

The HTTP server now supports terminating TLS directly via SSL environment variables — no reverse proxy required for simple self-hosted setups. Off by default; opt in by setting the relevant SSL env vars.

PR: https://github.com/roboflow/inference/pull/2308

---

# 🔒 Security & Hardening

* 🚫 **Block legacy fine-tuned SAM3 loads** — tighten the SAM3 model loading boundary so legacy fine-tuned artefacts that no longer match the supported package shape are rejected up front. [#2278](https://github.com/roboflow/inference/pull/2278)
* 📦 **OpenTelemetry stack bumped** to `1.41.x` / `0.62b1` to clear a transitive `protobuf` CVE and unlock alignment with the newer `inference-models` release. [#2335](https://github.com/roboflow/inference/pull/2335)
* 🛡️ **Additional security fixes** across the stack. [#2336](https://github.com/roboflow/inference/pull/2336)

---

# 🚧 Maintenance

* Align rfdetr preproc/postproc to rf-detr-internal by @leeclemnet in https://github.com/roboflow/inference/pull/2298
* Fix workflow block display names for acronym segments by @SolomonLake in https://github.com/roboflow/inference/pull/2320
* Fix owlv2 / rf-instant forward for best/default confidence modes by @leeclemnet in https://github.com/roboflow/inference/pull/2324
* Drop default/best confidence parametrization from OWLv2HF test by @leeclemnet in https://github.com/roboflow/inference/pull/2337
* Fix import for Gaze model by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2338
* Constraint `transformers` on Jetson with JP5 by @PawelPeczek-Roboflow in https://github.com/roboflow/inference/pull/2339
* Update serverless docs to v2 and prune deprecated pages by @Erol444 in https://github.com/roboflow/inference/pull/2311
* chore(docs): add GEO/SEO schema, robots.txt, git-revision-date plugin by @Borda in https://github.com/roboflow/inference/pull/2312

---

# 👋 New Contributors

* @Borda made their first contribution in https://github.com/roboflow/inference/pull/2312

**Full Changelog**: https://github.com/roboflow/inference/compare/v1.2.7...v1.2.8